The speed of the carriage held fairly steady between 2 and 3 centi- 

 meters per second and on the final pass a "standard" accumulation of 

 sand, about 2 centimeters high, was pushed along by the scraper. 

 Flexible plastic "wings" attached to the sides of the scraper pressed 

 against the channel walls and removed any "meniscus" of sand. The 

 motor-driven scraper could not reach within 3 or 4 centimeters of the 

 ends of the bed and these areas were leveled afterward by a hand- 

 operated scraper held against the floors of the spools. 



Pictures of the surface of the sand bed and of ripple profiles 

 were taken with a 35-millimeter Canon camera. For reasons of space 

 and parallax, each picture was made to include only one third of the 

 test section, so that a quasi-instantaneous record of an entire profile 

 required a quickly taken triad of pictures with the camera in three 

 successive positions. A clock which moved with the camera showed the 

 time in each picture. The camera could be placed at two levels, the 

 higher providing an oblique view of the surface and the lower a direct 

 side view of the profile. Various lighting was selected for camera 

 position and for low and moderate or large bed forms. Threads taped 

 to the tank wall served as gridlines, with vertical lines 10 centi- 

 meters apart and with horizontal lines at the level of the undisturbed 

 bed and 5 centimeters above and below. 



2. Ranges of Materials and Variables . 



Three sands were used in the experiments; grain-size distributions 

 (by sieve analysis at the Coastal Engineering Research Center (CERC) in 

 1977) are shown in Figure 7 and photos of the grains are shown in Figure 

 8. The sands are referred to by a value of Dg„ as 0.55, 0.21, and 0.18 

 millimeter. The three sands, with grain-size distributions, were sup- 

 plied by CERC. 'I'he 0. 21 -millimeter sand is the CERC Shore Processes Test 

 Basin (SPTB) sand after it had been further washed and strained at the 

 National Bureau of Standards (NBS) to remove the finest and coarsest 

 grains. Although the 0.21- and 0.18-millimeter sands differ little in 

 average grain diameters, the distribution of the 0.21-millimeter sand is 

 relatively broad and contains grains of a "medium" size. The specific 

 gravities of the three sands (measured at NBS) were found to lie in the 

 range 2.650 + 0.004, and all have been rounded to 2.65, the usual value 

 for quartz. 



In Tables A-1, A-2, and A-3 in Appendix A, the experiments are 

 numbered and listed chronologically with their conditions and with 

 selected observed quantities (one table for each sand) . The water tem- 

 perature remained fairly constant throughout the experiments, with 

 readings all within 4° Celsius and over half within 1° Celsius of the 

 average (21.2° Celsius). Therefore, v has been regarded as constant 

 (at 10"^ centimeter squared per second) and r takes on the constant 

 values: 52.39, 13.25, and 9.81 for the 0.55-, 0.21-, and 0.18-millimeter 

 sands, respectively. The independent variable a is conveniently 

 determined by N, which is simply the niomber of turns of a nut setting 

 the stroke of the pistons such that, 



a = 0.4576 N centimeters . (15) 



30 



